Limerick: Employers across this part of Ireland spent the winter enrolling staff into a retirement scheme most of those workers had never asked for, and almost none of them opted out. Ireland switched on auto-enrolment on 1 January 2026, the last country in western Europe to do it, and the early behaviour matches what behavioural economists predicted decades ago. People save when saving is the default.
Brussels has noticed. The Commission’s supplementary pensions package, published on 20 November 2025 as part of the Savings and Investments Union, urges member states to widen auto-enrolment, build pension tracking systems and publish national dashboards so citizens can see what they have accumulated. It sits alongside legislative proposals to revise the directive governing occupational pension funds and the regulation behind the pan-European personal pension product.
The legislative half has moved. The Council agreed its position on the personal pension product on 24 June 2026 and on the workplace pensions directive two days later, and it has published both mandates through the Council press service. Trilogues cannot start until Parliament settles its own position, and the economic affairs committee is still working. A workshop on auto-enrolment and supplementary pensions convenes in Brussels on 10 September, which tells you the Commission is still building consensus rather than banking it.
The instruments do not match the ambition
The problem is that the strongest lever in this package is the weakest instrument. Auto-enrolment arrives as a recommendation, and a recommendation binds nobody. Whether default enrolment produces real savings depends on the contribution rate, the employer match, the tax treatment and the quality of the default fund. Every one of those sits with national governments and, in most cases, with finance ministries defending revenue. The Union can describe good practice. It cannot set a contribution rate in Naples or a tax relief threshold in Warsaw.
The pan-European personal pension product shows what happens when the design misses. Launched with a one per cent annual fee cap that was meant to protect savers, it attracted almost no providers because the cap made distribution uneconomic. The Commission now proposes to remove it. That is an honest correction, and it also concedes that the original product failed on commercial grounds rather than on demand.
Whose money, and for what
The wider Savings and Investments Union argument runs like this. European households hold enormous balances in deposits earning little, European firms complain they cannot raise growth capital at home, and pensions could bridge the two if more people saved through funded schemes invested in European assets. The logic holds. The execution runs into a distributional question that the strategy documents tend to skip.
Funded pensions transfer risk to the saver. A defined benefit promise puts investment risk on the sponsor, while a defined contribution pot puts it on the household. Countries with mature funded systems learned that lesson through market cycles and built disclosure, guarantees and default-fund regulation around it. Countries starting now would be adopting the mechanism without the accumulated protections, and a recommendation is a thin vehicle for exporting those protections.
There is also a sequencing risk. The Commission wants pension money invested more heavily in European equity and infrastructure, and it is reviewing prudential rules partly with that in mind. Supervisors have to weigh that against their duty to members, whose interest is a good retirement rather than a well-capitalised continent. Those objectives usually align. When they do not, the directive being negotiated now decides which one wins, and that choice deserves more scrutiny than a technical file normally attracts.
Ireland’s opening months suggest the behavioural insight travels. The institutional question is different and harder. Europe can standardise what a pension statement looks like and how a fund explains underperformance. It cannot legislate the contribution that makes the statement worth reading, and until member states make that choice themselves, the dashboards will show a great deal of clearly presented emptiness.
